Wideband speech encoders have been introduced over the last few years in order to allow a vast improvement in the quality and intelligibility of voice, or speech, communicated between wired or wireless communication devices. Typically, the wideband speech vocoders include voice encoders that are operable in a number of different encoder modes, and selection of an encoder mode for a given call is subject to a set of rules that, in part, depend on the standards relevant to the particular communication technology. In general, the encoder mode for a given call may be selected such that the encoder mode is either a wideband encoder mode or a narrowband encoder mode, where the selection is based on signaling protocols exercised at the time that the call is setup or, alternatively, based on in-band signaling exchanged via bearer signal packets. As an example, the Third Generation Partnership Project 2 (3GPP2) Enhanced Variable Rate Codec-Narrowband-Wideband (EVRC-NW) vocoder includes a wideband encoder mode and multiple narrowband encoder modes. Currently, the encoder mode of the EVRC-NW vocoder is determined based on the transmission of a request for a preferred encoder mode.
More specifically, when a call is established between two mobile terminals (mobile terminal A and mobile terminal B) equipped with EVRC-NW vocoders, a bearer path is established between corresponding control nodes (e.g., base stations or mobile switching centers) in the cellular communication system over a core network. This bearer path is used to carry encoded speech between the control nodes of the mobile terminals. For Code Division Multiple Access (CDMA) cellular communication systems and the EVRC family of vocoders, the bearer path over a core network is provided via Real Time Protocol (RTP) packets. As illustrated in FIG. 1, encoder mode requests are included in a dedicated field, referred to as an MMM field, in the payload headers of the RTP packets transmitted over the core network to carry encoded speech, as specified in, for example, the Internet Engineering Task Force (IETF) Request for Comment (RFC) 4788.
Notably, an RTP packet has a header and a payload. The RTP packet header carries information such as time stamp, sequence number, etc. The RTP packet header information is generic and typically independent of the payload. The payload of the RTP packet can be any media payload such as video, audio, text, etc. Here, the payload of the RTP packet is used to carry encoded speech over the core network. Specifically, the payload carries the EVRC-NW encoded speech. The EVRC-NW payload has its own header defined, which is different from the RTP packet header. The EVRC-NW payload header includes the MMM field as illustrated in FIG. 1.
One issue with the current encoder selection scheme for EVRC-NW vocoders is that the request for the preferred encoder mode is sent with no awareness of the current capability of the voice encoder of the far-end vocoder to accommodate the request. More specifically, for the call between mobile terminal A and mobile terminal B, the control node for mobile terminal A selects and requests a preferred encoder mode for the voice encoder of mobile terminal B with no awareness of the current capability of the voice encoder of mobile terminal B or limitations that may be imposed due to the call configuration. Likewise, the control node for mobile terminal B selects and requests a preferred encoder mode for the voice encoder of mobile terminal A with no awareness of the current capability of the voice encoder of mobile terminal A. As such, if for example mobile terminal A is capable of receiving and decoding wideband speech, the control node of mobile terminal A will continually request the wideband encoder mode regardless of whether the voice encoder of mobile terminal B or the call configuration can support operation in the wideband encoder mode. If the voice encoder of mobile terminal B is incapable of wideband encoding (e.g., because the communication node does not support the wideband mode, because the communication node supports the wideband encoder mode but is currently not capable of wideband encoding because the communication node is operating in a sector that does not support wideband operation, because wideband encoding is not permitted by local operator policy, or the like), the voice encoder of mobile terminal B will use some default narrowband encoder mode which may not be a preferred narrowband encoder mode of mobile terminal A. In other words, mobile terminal A will receive speech encoded according to one of the narrowband encoder modes without being aware that the voice encoder of mobile terminal B is incapable of wideband encoding and, consequently, without being able to signal a preference for a preferred narrowband encoder mode.
This problem is made worse if the wireless network operator's service policy is to proffer the highest priority to wideband operation (e.g., to gain maximum customer satisfaction especially during an introductory phase of wideband speech). In this case, there is no choice other than to keep requesting the wideband encoder mode since the reception of narrowband speech does not preclude events such as handover to a territory that supports wideband speech, which might suddenly make it possible for the voice encoder of mobile terminal B to commence transmission of wideband speech. In light of the discussion above, there is a need for an improved encoder mode selection scheme for wideband vocoders and, in particular, EVRC-NW vocoders.